On 2026-07-27 13:22, Jakub Jelinek wrote:
On Mon, Jul 27, 2026 at 09:58:50AM +0000, Torbjörn Svensson via Sourceware
Forge wrote:
Currently, it's assumed that arm_fp16_alternative_ok is enough to run
the tests, but arm_fp16_alternative_ok allows multilib override. While
the test might link without trouble, execution will likely fail.
Add arm_fp16_alternative_hw to explicitly test that target can execute
fp16 instructions in alternative form.
The last change in sourcebuild.texi uses @code(...) which makeinfo
errors on:
sourcebuild.texi:2122: @code expected braces
sourcebuild.texi:2131: @code expected braces
Fixed thusly, tested on x86_64-linux, committed to trunk as obvious.
Ahh. I should have tested the suggestion rather than just trusting it.
I've also picked the change to releases/gcc-15 that I picked the breaking
change to.
Sorry for the problem this might have caused.
Kind regards,
Torbjörn
2026-07-27 Jakub Jelinek <[email protected]>
* doc/sourcebuild.texi (arm_neon_fp16_hw, arm_fp16_alternative_hw):
Use @code{...} instead of @code(...).
--- gcc/doc/sourcebuild.texi.jj 2026-07-27 12:45:27.493830709 +0200
+++ gcc/doc/sourcebuild.texi 2026-07-27 13:17:00.955318019 +0200
@@ -2119,7 +2119,7 @@ options, including @code{-mfp16-format=i
@item arm_neon_fp16_hw
Test system supports executing Neon half-precision float instructions.
-(Implies @code(arm_neon_fp16_ok).)
+(Implies @code{arm_neon_fp16_ok}.)
@item arm_fp16_alternative_ok
ARM target supports the ARM FP16 alternative format. Some multilibs
@@ -2128,7 +2128,7 @@ may be incompatible with the options nee
@item arm_fp16_alternative_hw
Test system supports executing VFP half-precision floating-point
instructions in the ARM FP16 alternative format. (Implies
-@code(arm_fp16_alternative_ok).)
+@code{arm_fp16_alternative_ok}.)
@item arm_fp16_none_ok
ARM target supports specifying none as the ARM FP16 format.
Jakub
